Deep Learning-Driven Big Data Analytics in the Agri-Food Sector: A Systematic Review

The agri-food sector is a key promoter of global food security and economic stability. Post-harvest loss, wasteful resources, climatic uncertainty, and sustainability requirements are risks that require intelligent, data-driven solutions. Evolving opportunities in Big Data Analytics (BDA) and Deep Learning (DL) have been found to have massive potential to revamp agricultural models, food processing, and preservation networks. This systematic review covers 20 studies between 2022 and 2025 based on deep learning approaches through big data predictive analytics, crop monitoring, yield estimation, quality inspection, and smart preservation methods. The review lists the application of convolutional, recurrent, and attention-based neural networks on different agricultural datasets for better decision-making, process automation, and optimization of efficiency. The review also covers the issues of data management such as heterogeneity, scalability, interpretability, and integration throughout the agri-food value chain. Summarizing the latest success, the review presents its position on how big data systems powered by DL can transform the agri-food industry for dedication to resilience, accuracy, and sustainability in the global food system.